Nandan Kumar Jha

34 papers A* 3A 2Misc 2Journal 24Unranked 3

Year	Rank	Type	Title / Venue / Authors
2026	J	`jnl`	NerVE: Nonlinear Eigenspectrum Dynamics in LLM Feed-Forward Networks. CoRR Nandan Kumar Jha, Brandon Reagen
2025	J	`jnl`	A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention. CoRR Nandan Kumar Jha, Brandon Reagen
2025	J	`jnl`	Entropy-Guided Attention for Private LLMs. CoRR Nandan Kumar Jha, Brandon Reagen
2025	A	`conf`	Network and Compiler Optimizations for Efficient Linear Algebra Kernels in Private Transformer Inference (Invited Paper). ICCAD Karthik Garimella, Negar Neda, Austin Ebel, Nandan Kumar Jha, Brandon Reagen
2025	J	`jnl`	Network and Compiler Optimizations for Efficient Linear Algebra Kernels in Private Transformer Inference. CoRR Karthik Garimella, Negar Neda, Austin Ebel, Nandan Kumar Jha, Brandon Reagen
2025	J	`jnl`	Spectral Scaling Laws in Language Models: How Effectively Do Feed-Forward Networks Use Their Latent Space? CoRR Nandan Kumar Jha, Brandon Reagen
2025	A*	`conf`	Spectral Scaling Laws in Language Models: emphHow Effectively Do Feed-Forward Networks Use Their Latent Space? EMNLP Nandan Kumar Jha, Brandon Reagen
2024	J	`jnl`	AERO: Softmax-Only LLMs for Efficient Private Inference. CoRR Nandan Kumar Jha, Brandon Reagen
2024	J	`jnl`	DeepReShape: Redesigning Neural Networks for Efficient Private Inference. Trans. Mach. Learn. Res. Nandan Kumar Jha, Brandon Reagen
2024	J	`jnl`	ReLU's Revival: On the Entropic Overload in Normalization-Free Large Language Models. CoRR Nandan Kumar Jha, Brandon Reagen
2024	J	`jnl`	TruncFormer: Private LLM Inference Using Only Truncations. CoRR Patrick Yubeaton, Jianqiao Mo, Karthik Garimella, Nandan Kumar Jha, Brandon Reagen, Chinmay Hegde, Siddharth Garg
2023	—	`conf`	Characterizing and Optimizing End-to-End Systems for Private Inference. ASPLOS (3) Karthik Garimella, Zahra Ghodsi, Nandan Kumar Jha, Siddharth Garg, Brandon Reagen
2023	J	`jnl`	DeepReShape: Redesigning Neural Networks for Efficient Private Inference. CoRR Nandan Kumar Jha, Brandon Reagen
2022	J	`jnl`	Characterizing and Optimizing End-to-End Systems for Private Inference. CoRR Karthik Garimella, Zahra Ghodsi, Nandan Kumar Jha, Siddharth Garg, Brandon Reagen
2021	A*	`conf`	Circa: Stochastic ReLUs for Private Deep Learning. NeurIPS Zahra Ghodsi, Nandan Kumar Jha, Brandon Reagen, Siddharth Garg
2021	J	`jnl`	Circa: Stochastic ReLUs for Private Deep Learning. CoRR Zahra Ghodsi, Nandan Kumar Jha, Brandon Reagen, Siddharth Garg
2021	J	`jnl`	CryptoNite: Revealing the Pitfalls of End-to-End Private Inference at Scale. CoRR Karthik Garimella, Nandan Kumar Jha, Zahra Ghodsi, Siddharth Garg, Brandon Reagen
2021	A*	`conf`	DeepReDuce: ReLU Reduction for Fast Private Inference. ICML Nandan Kumar Jha, Zahra Ghodsi, Siddharth Garg, Brandon Reagen
2021	J	`jnl`	DeepReDuce: ReLU Reduction for Fast Private Inference. CoRR Nandan Kumar Jha, Zahra Ghodsi, Siddharth Garg, Brandon Reagen
2021	—	`conf`	Digital Storytelling: The Integration of Intangible and Tangible Heritage in the City of Surat, India. HCI (33) Chika Udeaja, Lukman E. Mansuri, Busisiwe Chikomborero Ncube Makore, Kwasi Gyau Baffour Awuah, Dilip A. Patel, Claudia Trillo, Nandan Kumar Jha
2021	J	`jnl`	Modeling Data Reuse in Deep Neural Networks by Taking Data-Types into Cognizance. IEEE Trans. Computers Nandan Kumar Jha, Sparsh Mittal
2021	J	`jnl`	Sisyphus: A Cautionary Tale of Using Low-Degree Polynomial Activations in Privacy-Preserving Deep Learning. CoRR Karthik Garimella, Nandan Kumar Jha, Brandon Reagen
2020	—	`conf`	DRACO: Co-Optimizing Hardware Utilization, and Performance of DNNs on Systolic Accelerator. ISVLSI Nandan Kumar Jha, Shreyas Ravishankar, Sparsh Mittal, Arvind Kaushik, Dipan Mandal, Mahesh Chandra
2020	J	`jnl`	DRACO: Co-Optimizing Hardware Utilization, and Performance of DNNs on Systolic Accelerator. CoRR Nandan Kumar Jha, Shreyas Ravishankar, Sparsh Mittal, Arvind Kaushik, Dipan Mandal, Mahesh Chandra
2020	J	`jnl`	DeepPeep: Exploiting Design Ramifications to Decipher the Architecture of Compact DNNs. CoRR Nandan Kumar Jha, Sparsh Mittal, Binod Kumar, Govardhan Mattela
2020	J	`jnl`	DeepPeep: Exploiting Design Ramifications to Decipher the Architecture of Compact DNNs. ACM J. Emerg. Technol. Comput. Syst. Nandan Kumar Jha, Sparsh Mittal, Binod Kumar, Govardhan Mattela
2020	Misc	`conf`	E2GC: Energy-efficient Group Convolution in Deep Neural Networks. VLSID Nandan Kumar Jha, Rajat Saini, Subhrajit Nag, Sparsh Mittal
2020	J	`jnl`	E2GC: Energy-efficient Group Convolution in Deep Neural Networks. CoRR Nandan Kumar Jha, Rajat Saini, Subhrajit Nag, Sparsh Mittal
2020	J	`jnl`	Modeling Data Reuse in Deep Neural Networks by Taking Data-Types into Cognizance. CoRR Nandan Kumar Jha, Sparsh Mittal
2020	J	`jnl`	On the Demystification of Knowledge Distillation: A Residual Network Perspective. CoRR Nandan Kumar Jha, Rajat Saini, Sparsh Mittal
2020	J	`jnl`	The Ramifications of Making Deep Neural Networks Compact. CoRR Nandan Kumar Jha, Sparsh Mittal, Govardhan Mattela
2020	A	`conf`	ULSAM: Ultra-Lightweight Subspace Attention Module for Compact Convolutional Neural Networks. WACV Rajat Saini, Nandan Kumar Jha, Bedanta Das, Sparsh Mittal, C. Krishna Mohan
2020	J	`jnl`	ULSAM: Ultra-Lightweight Subspace Attention Module for Compact Convolutional Neural Networks. CoRR Rajat Saini, Nandan Kumar Jha, Bedanta Das, Sparsh Mittal, C. Krishna Mohan
2019	Misc	`conf`	The Ramifications of Making Deep Neural Networks Compact. VLSID Nandan Kumar Jha, Sparsh Mittal, Govardhan Mattela